Biography

I received my Ph.D. from Cornell University in 2024, advised by Prof. Christina Delimitrou. My research focuses on improving the performance and efficiency of cloud and machine learning systems. I was a research intern at Google Brain in 2021 and at Meta in 2022 and 2024.

Before joining Cornell, I obtained my B.S. degree from Shanghai Jiao Tong University.

I am currently a software engineer at Google DeepMind, working on Gemini serving performance.

Education

Ph.D. in Electrical and Computer Engineering, Cornell University, 2024

Bachelor in Computer Science, Shanghai Jiao Tong University, 2019

Experience

Research intern: 2024
Meta, mentored by Krishna Malladi
Project: Hardware architecture design for ML inference

Research intern: 2022
Meta, mentored by Wenyin Fu
Project: Production AI benchmarks generation

Research intern: 2021
Google Brain, mentored by Martin Maas
Project: Large-scale distributed ML workloads training and scheduling

Selected Publications

Mingyu Liang, Hiwot Tadese Kassa, Wenyin Fu, Brian Coutinho, Louis Feng, and Christina Delimitrou. “Lumos: Efficient Performance Modeling and Estimation for Large-scale LLM Training”. In Proceedings of the 8th Annual Conference on Machine Learning and Systems (MLSys’25). [pdf]

Mingyu Liang, Hiwot Tadese Kassa, Wenyin Fu, Brian Coutinho, Louis Feng, and Christina Delimitrou. “Fine-grained Trace-driven Performance Modeling and Simulation for Large-scale ML Training”. In the Workshop on Machine Learning for Computer Architecture and Systems (MLArchSys’24). [pdf]

Mingyu Liang, Wenyin Fu, Louis Feng, Zhongyi Lin, Pavani Panakanti, Shengbao Zheng, Srinivas Sridharan, and Christina Delimitrou. “Mystique: Enabling Accurate and Scalable Generation of Production AI Benchmarks”. In 50th International Symposium on Computer Architecture (ISCA’23). [pdf]

Mingyu Liang*, Yu Gan*, Yueying Li, Carlos Torres, Abhishek Dhanotia, Mahesh Ketkar, and Christina Delimitrou. “Ditto: End-to-End Application Cloning for Networked Cloud Services”. In 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS’23). Selected in IEEE Micro's Top Picks special issue of “most significant papers in computer architecture based on novelty and long-term impact” for 2023. [pdf]

Yu Gan, Mingyu Liang, Sundar Dev, David Lo, and Christina Delimitrou. “Sage: Practical & Scalable ML-Driven Performance Debugging in Microservices”. In 26th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS’21). Selected in IEEE Micro's Top Picks special issue of “most significant papers in computer architecture based on novelty and long-term impact” for 2021. [pdf]

Honors & Awards

ML and Systems Rising Stars, ML Commons, 2023

IEEE Micro’s Top Picks, for the paper “Ditto: End-to-End Application Cloning for Networked Cloud Services”, 2023

IEEE Micro’s Top Picks, for the paper “Sage: Practical & Scalable ML-Driven Performance Debugging in Microservices”, 2021

Cornell Graduate Fellowship, 2019

Zhiyuan College Honors Scholarship, 2016, 2017

Teaching Experience

Computer Architecture (ECE 4750), Fall 2024

Embedding Systems (ECE 3140), Spring 2023

Datacenter Computing (ECE 5710), Spring 2021

Contact

Email: ml2585@cornell.edu